Could 'Menlo' signal a change in Microsoft's mobile strategy?

Microsoft "Menlo" is a hush-hush research project that is focused on replacing Windows Compact Edition (CE) with Windows NT, according to my sources. "Experiment 19" is a graphics platform researchers are building to complement Menlo, my sources have told me. I don't have a lot of details about these Research efforts beyond that -- so far, that is....
Written by Mary Jo Foley, Senior Contributing Editor

I've been digging for a few months for information about two new Microsoft codenames and have come up with few clues. But I've decided to share what I have been able to surface in the hopes that others who might have additional pieces to share might help fill things in.

Microsoft "Menlo" is a hush-hush research project which seems to be focused on replacing Windows CE (known officially these days as Windows Embedded Compact) Compact Edition (CE) with Windows NT inside of mobile devices, according to my sources. "Experiment 19" is a graphics platform researchers are building to complement Menlo, my sources have told me.

The lead for Menlo is Galen Hunt, the Microsoft researcher who spearheaded Microsoft's Singularity research project, as he acknowledges on his bio page on the Microsoft Research Web site. (Singularity is a microkernel-based operating system developed by Microsoft Research, and supposedly the starting point for Microsoft's Midori effort.) Researcher Ruben Olinsky also is on the Menlo team, focusing specifically on "Experiment 19," as is/was Kerry Hammil, a (former) Microsoft Researcher who has worked on 2D/3D graphics programming interfaces for Windows.

Update: A reader sent in the following tidbit. Hunt's LinkedIn profile includes this information:

"Recruited researchers and engineers to build the MSR Operating Systems research group. Co-lead two of the largest cross-group research projects in MSR’s history: the Menlo and Singularity projects. Menlo combined OS, UX, and applications research to explore the future of computing when mobiles becomes users primary PCs. Singularity combined language, tools, and OS research to determine how to build more reliable systems and built a new OS from scratch in managed code. Managed cross-division relationships with product groups and relationships with key hardware partners, negotiated production of custom phone prototypes, designed and implemented major kernel and OS components, implemented runtime and compiler features, and wrote code for demos."

The few references to "Menlo" that I've found on the Microsoft Research page are brief and generic. From Olinsky's bio: "I work in Microsoft Research's Operating Systems Group as a research developer.  I'm currently working on the Menlo project and spending quite a bit of time tinkering in the mobile computing space." (Like Hammil, Olinsky has worked on technology that has found its way into Windows Vista/Windows 7.)

Here's another Menlo reference, courtesy of the bio page from another Microsoft Researcher, A.J. Brush, regarding a paper that will be published in September 2010:

"User Experiences with Activity-Based Navigation on Mobile Devices A.J. Bernheim Brush, Amy Karlson, James Scott and Menlo team members, To appear in Mobile HCI 2010"

Based on these references, it looks like Menlo definitely has something to do with Microsoft's future strategy in the mobile space. But exactly why the company may be looking into replacing the CE core -- which is at the heart of many third-party devices, as well as Microsoft products including Windows Mobile, Windows Phone OS 7.0 and the Zune operating system -- is still not 100 percent clear.

"(Windows) CE is actually not their problem and has never been the problem," said one of my doubting sources. "They are investing in Silverlight as the UI and runtime for third party applications and not in the Win32CE API. They want developers to use that framework instead of the native code framework."

When I described to Directions on Microsoft analyst Michael Cherry about what I'd heard so far about Menlo, he called the concept a potential "frankenOS." He noted that Apple has begun to redefine the concept of the tablet OS (with the iPad), convincing customers and industry watchers that the best platform to power these kinds of devices isn't a traditional client OS like Windows. Cherry said he was doubtful Microsoft would need to go to all the trouble of replacing CE with NT.

Instead, Cherry suggested two alternative courses of action:

"Microsoft could follow Apple’s lead and build a similar tablet on CE or Windows Phone 7, and have developers create CE and Web based applications for it....Mostly, it would be a rival to the iPad with similar battery life attributes.  Microsoft could even add some USB ports just to differentiate the connectability.  They could also open the carrier access.

"Second, Microsoft could build a tablet on Windows 7 Standard Embedded, and really be ruthless in removing Windows 7 components that do not make sense in the 'new' tablet scenarios. I have not seen any attempt to use what they have to create a truly “designed for tablet” version of OS (key here is instead of adding tablet extensions you remove non-tablet components of the base OS)."

Here's my thinking: If Microsoft replaced the CE kernel with an NT one, Microsoft might have a more compelling "write once/run anywhere story." What if you could make some relatively minor tweaks to your existing Windows client app to get them to run on a Menlo-based Windows Phone or Menlo-based tablet/slate? (It would be the reverse of what Apple is doing, by enabling developers to move their iPhone apps to the iPad with relatively little work.)

Because Microsoft has a far bigger pool of Windows apps than Windows Mobile/Windows Phone apps, it would make sense for the Redmondians to be trying to find a way to enable that base of business/consumer apps to run on Windows-centric phones and tablets.

I'm also wondering whether Menlo may be related to Microsoft's rumored work to port Windows to the ARM platform. There was a project inside Microsoft codenamed "LongARM," my sources have said, via which Microsoft was working on porting Windows Vista to ARM. So far, that port has not materialized. Windows CE runs fine on ARM. If Menlo means Microsoft can swap out CE for NT, would it mean that NT-based operating systems -- like Windows -- could run everywhere CE did?

Menlo, like "Newport," a Microsoft Research app for mobile collaboration, seems to be part of Microsoft's grander strategy of making three-screen-and-a-cloud computing simpler and more ubiquitous. If Menlo really is about enabling mobile devices to run the same "core" as Windows itself, I'd think the PC-phone synchronization part of the three-screens strategy would get a lot easier.

In case you haven't gathered by now, Microsoft Research is not commenting about Menlo. I'm figuring my chances of someone leaving a Winpad running Menlo on a barstool at my local watering hole to be slim to none. (And even if someone did, I don't have a budget to pay a tipster a finder's fee for it.) Given Boy Genuis' recent estimate that 70 percent of tips these days are plants/intentional leaks -- a number I find totally believable -- I decided I'd beat the leakers to the punch.

Can anyone help figure out the Microsoft Menlo mystery? Feel free to chime in below with guesses (educated and otherwise). Or if you'd rather comment privately (and confidentially), email me using the link at the bottom of this site.

Editorial standards